Pancreatic ductal adenocarcinoma (PDA) is the fourth leading cause of cancer death. The overall goal of these studies is to gain a deeper understanding of the biological functions of pancreatic stellate cells (PSCs), a small subpopulation of pancreatic stromal cells, in the progression of pancreatic ductal adenocarcinoma (PDA), and to define molecular mechanisms underlying stellate cell activation and their interactions with pancreatic tumor cells. Reciprocal signaling via cytokines and extracellular matrix proteins between pancreatic cancer cells (PCCs) and PSCs leads to sustained activation of PSCs and also promotes tumor progression and metastasis. To investigate paracrine interactions between human PSCs and human PCCs, MS was used to define the secretomes of the two cell types. The LIF IL6 family cytokine was found to be the major factor secreted by PSCs that stimulates PCCs, and, conversely, PDGF was identified as a major PCC-secreted factor that stimulates PSCs, setting up a reciprocal stimulatory loop. The expression of LIF in human PDA tumor samples will be assessed in order to validate cell culture results, and the role of LIF in PDA will be assessed using orthotopic xenografts with parental or LIF receptor-depleted PCCs co-injected with PSCs. The PDGF-induced pathway involving the UBASH3B/STS1 adaptor protein that induces LIF expression in PSCs will be investigated. A subset of human PCC lines has been found to be resistant to MEK inhibitors, even though they express mutant KRAS that drives activation of the ERK MAP kinase pathway through MEK. MEK inhibitor sensitive PCC lines are rendered resistant to MEK inhibition by treatment with conditioned medium from resistant PCC lines. By MS analysis and partial characterization of the resistance factor, GDF-15, a variant TGF family member, was identified as a candidate, and the possible role of GDF15 in MEK inhibitor resistance will be evaluated by reverse genetic manipulation and use of neutralizing antibodies. Alternatively, the resistance factor will be purified by conventional means. To study the relationship between PDA progression and PSC activation in vivo, a genetically engineered mouse model, SLT, has been made, in which luciferase (Luc) and reverse tetracycline transactivator (rtTA) are specifically expressed in activated stellate cells, enabling both bioluminescent labeling and doxycycline-inducible manipulation of activated PSCs in the pancreas. By crossing SLT mice with KPflC PDA model mice, PSC activation in the course of PDA progression will be studied. SLT;tetO-DTA;KPflC mice, in which diphtheria toxin A chain (DTA) is inducible by doxycycline feeding, will be generated and used to evaluate the consequences of PSC inactivation by DTA on PDA progression at defined times after the initiation of tumorigenesis. Finally, KPflC;LIFRfl/fl mice wll be generated, and comparison with KPflC mice will be used to evaluate how deficiency of LIFR signaling specifically in tumor cells affects PDA progression. The efficacy of LIF blockade for PDA therapy will be assessed in the KPflC mouse model by administration of soluble dominant-negative Fc-LIFR ECD fusion protein.